problem with run_inference.py about llama2lang

[ErikTromp]

That is expected! The script now runs interactively so you can type your input once you see that prompt. Then it will generate a response after which you can iterate.

Let me know if that works for you

[elBlacksmith]

That is expected! The script now runs interactively so you can type your input once you see that prompt. Then it will generate a response after which you can iterate.

Let me know if that works for you

Thanks for the quick reply.
The problem was that Google Colab doesn´t show clearly the input form, so I missed it and I was confused.

BTW, what software do you recommend to execute the qlora?

[ErikTromp]

So actually Colab does show the input and allows you to write something but you have to click right next to the prompt message to make it visible.

[elBlacksmith]

So actually Colab does show the input and allows you to write something but you have to click right next to the prompt message to make it visible.

Well, that´s how (bad) Google Colab is designed.
Let me show you.

This is the output:

And this is shown when you click on it (If you know exactly where to click):